BM25 search across all repos in a group, merged via Reciprocal Rank Fusion (RRF). Returns a unified ranked list plus per-repo breakdown.
AI agents call group_query to retrieve information from code-intel MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is a search/query tool that reads data across repositories and returns results. It has no capability to modify, execute, delete, or create data. The blast radius of misuse is minimal — an attacker could only discover information already present in the codebase, not alter systems or cause harm. BM25 is a standard information retrieval algorithm with no side effects.
From the tool's definition Tool performs 'BM25 search' and 'Returns a unified ranked list' — it retrieves and queries data with no mutation or side effects. The description contains no language suggesting code execution, data modification, deletion, or financial operations.
Attacks that exploit this kind of access
BM25 search across all repos in a group, merged via Reciprocal Rank Fusion (RRF). Returns a unified ranked list plus per-repo breakdown. It is categorised as a Read tool in the code-intel MCP Server MCP Server, which means it retrieves data without modifying state.
Register the code-intel MCP Server MCP server in PolicyLayer and add a rule for group_query: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches code-intel MCP Server. Nothing to install.
group_query is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the group_query rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for group_query. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
group_query is provided by the code-intel MCP Server MCP server (vohongtho/code-intel-platform). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →